| Index | Recent Threads | Unanswered Threads | Who's Active | Guidelines | Search |
 |
World Community Grid Forums
Category: Active Research
Forum: OpenPandemics - COVID-19 Project
Thread: OpenPandemics - COVID-19 is Now Live |
| No member browsing this thread |
|
Thread Status: Active Thread Type: Sticky Thread Total posts in this thread: 282
|
|
| Author |
|
|
Falconet
Master Cruncher Portugal Joined: Mar 9, 2009 Post Count: 3315 Status: Offline Project Badges: 
|
Simplex0,
----------------------------------------We're not really folding proteins but rather docking molecules against a protein target. So basically, we're trying to find if a certain molecule manages to fit in some part of a protein and prevent that protein from carrying out its function (replication, etc). Each task you run is trying to dock several molecules (jobs as they are called in the result log). Each of those jobs is one docking of a molecule against 1 protein target and each task usually has several jobs. In the case of CPU tasks, there are usually only a few jobs (normally 2 jobs but I have seen reports of up to 5) while GPU tasks have a lot more (highest I've seen was 68 jobs during BETA but there are also lower numbers such as 9 jobs). GPU tasks pack a lot more work than a CPU task but run a lot faster. So, if you want to find out how much time it takes to finish one docking, go to the result log of a completed CPU/GPU OPN task, check the time it started the job and then check the time when it finished the job. See this post on the BETA thread for an example of what I mean https://www.worldcommunitygrid.org/forums/wcg...ad,43290_offset,15#653833 As you can see in that post: INFO:[09:20:40] Start AutoDock for ZINC000923382694-ACR2.22_RX1--fr2266benz_001--CYS114.dpf(Job #0)... OpenCL device: Ellesmere INFO:[09:22:18] End AutoDock... This means it started docking against a protein target using ZINC000923382694, which is a molecule from the ZINC Library (you cane even go there and search for all these codes and see the molecules you worked on). It took less than 2 minutes on the GPU to dock the molecule against the protein target. I hope this is what you asked about.  - AMD Ryzen 5 1600AF 6C/12T 3.2 GHz - 85W - AMD Ryzen 5 2500U 4C/8T 2.0 GHz - 28W - AMD Ryzen 7 7730U 8C/16T 3.0 GHz |
||
|
|
cpalmer
Cruncher England Joined: Feb 14, 2021 Post Count: 16 Status: Offline Project Badges: 
|
Just noticed that other projects' cross-project stats, and sites like BoincStats, only seem to be showing my WCG CPU points, and not the GPU ones. At least that's what I think is happening. It certainly doesn't match up with what's shown on the WCG site...
|
||
|
|
fufu
Cruncher Joined: Mar 20, 2006 Post Count: 46 Status: Offline Project Badges: 
|
WCG points != Boinc points. This has nothing to do with just CPU points.
See here: https://www.worldcommunitygrid.org/help/viewTopic.do?shortName=points |
||
|
|
Simplex0
Advanced Cruncher Sweden Joined: Aug 14, 2008 Post Count: 83 Status: Offline Project Badges:
|
Simplex0, We're not really folding proteins but rather docking molecules against a protein target. So basically, we're trying to find if a certain molecule manages to fit in some part of a protein and prevent that protein from carrying out its function (replication, etc). Each task you run is trying to dock several molecules (jobs as they are called in the result log). Each of those jobs is one docking of a molecule against 1 protein target and each task usually has several jobs. In the case of CPU tasks, there are usually only a few jobs (normally 2 jobs but I have seen reports of up to 5) while GPU tasks have a lot more (highest I've seen was 68 jobs during BETA but there are also lower numbers such as 9 jobs). GPU tasks pack a lot more work than a CPU task but run a lot faster. So, if you want to find out how much time it takes to finish one docking, go to the result log of a completed CPU/GPU OPN task, check the time it started the job and then check the time when it finished the job. See this post on the BETA thread for an example of what I mean https://www.worldcommunitygrid.org/forums/wcg...ad,43290_offset,15#653833 As you can see in that post: INFO:[09:20:40] Start AutoDock for ZINC000923382694-ACR2.22_RX1--fr2266benz_001--CYS114.dpf(Job #0)... OpenCL device: Ellesmere INFO:[09:22:18] End AutoDock... This means it started docking against a protein target using ZINC000923382694, which is a molecule from the ZINC Library (you cane even go there and search for all these codes and see the molecules you worked on). It took less than 2 minutes on the GPU to dock the molecule against the protein target. I hope this is what you asked about. Tank you Falconet for taking the time and for a good explanation on how it works and on what is being done while processing the work units, I will follow your link and dig more in to how this process works. |
|||
|
||||
|
pwhidden
Cruncher USA Joined: Nov 17, 2004 Post Count: 32 Status: Offline Project Badges:
|
Well, I just did it. Ordered a new computer with a good graphics card just to crunch. i7-10700, GeForce GTX 1660Ti. I certainly hope WCG and the OpenPandemics team appreciate my financial contribution. (I am sure they do.)
----------------------------------------Now, if only the African Rainfall project could use the GPU! Uplinger and Keith: Thanks for all of your posts. (Also a member of the OFA!)  |
||
|
|
Stevie G
Cruncher United States Joined: Apr 10, 2020 Post Count: 24 Status: Offline Project Badges: 
|
On April 13, 2021, after a prolonged drought, my computer downloaded 24 WCG tasks. However, of those, 19 ended in errors.
OPNG_ 0001407_ 00027_ 0-- Microsoft Windows 7 Home Premium x64 Edition, Service Pack 1, (06.01.7601.00) 728 Error 4/13/21 17:40:09 4/15/21 18:10:48 0.02 1.1 / 0.0 Nineteen error messages like this one. Also, the computer was crunching these tasks very slowly, taking as long as an entire day, where it normally takes 2-3 hours for one. Finally, the time remaining was going backwards, getting longer by the second, while the elapsed time was ticking away. Very weird behavior, I'd say. Why was this happening? It seems to have returned to normal operation today. Thanks for any insight. S.Gaber |
||
|
|
shiftshift
Cruncher Australia Joined: Dec 13, 2015 Post Count: 3 Status: Offline Project Badges: 
|
I would guess that you are used to CPU tasks (OPN1). These tasks are GPU tasks (OPNG) and based on the errors and length of time your GPU isn't compatible. That is my guess.
---------------------------------------- |
||
|
|
Stevie G
Cruncher United States Joined: Apr 10, 2020 Post Count: 24 Status: Offline Project Badges:
|
shiftshift:
Thanks for that possible explanation. So if my GPU is not compatible, leading to prolonged completion times and errors, will BOINC automatically make an adjustment? Also, I have not received any new WCG tasks for a few days. Is there a general slowdown or do I have a wrong setting somewhere? S.Gaber |
||
|
|
ca05065
Senior Cruncher Joined: Dec 4, 2007 Post Count: 328 Status: Offline Project Badges: 
|
I am only asking for Open Pandemics work units at the moment. Yesterday I processed 208 work units (a mixture of OPN1 and OPNG) and I have my requested 64 work units either running or ready to start.
|
||
|
|
Stevie G
Cruncher United States Joined: Apr 10, 2020 Post Count: 24 Status: Offline Project Badges:
|
Shiftshift:
Now my computer shows it's running one WCG task and one Asteroids task. The WCG task says it has 00:23:56 minutes elapsed, counting down to 0:5:11:17 remaining. But the progress has remained at 5.882% without changing in the past 20 minutes. I suspect it will run down the 5+ hours remaining till it gets to zero, then will report it as an error while computing. So if this behavior is because CGU tasks are not compatible with my computer, how do I specify NO CGU tasks without it affecting WCG progress or any of my other BOINC projects? Thanks. S.Gaber |
||
|
|
|